INTRODUCTION:
Matin Movassate is the CEO of Heap, a tool that allows for instant, retroactive analysis of user actions in web or iOS apps. Hear what Matin has to say about who can benefit from Heap, how it compares to other tools on the market and what’s next for his growing company.
KEY LEARNINGS:
1. The process of data analysis can be very convoluted; Heap eliminates bottlenecks to data and bring insights to more people, like marketing, sales and product teams.
2. Heap automatically captures every user interaction, including clicks, form submissions, page-views, input changes, etc. to provide an interface for organizing and making sense of user behavior after the fact.
3. It’s important to focus on having the right data, which is the hardest part of analysis. It doesn’t matter how pretty your tool is, or how fast or advanced it is; if you don’t have the right data, your analysis is fundamentally hamstrung.
4. Heap relies on its customers to inform their future vision rather than watching every move of its competitors.
5. As the analytics field evolves, tools are going to start integrating with each other better. Heap hopes to build better integration with other analytics tools in 2016.
6. One of the best things an entrepreneur can do is to “sell” things before they are actually built. It saves a lot of time and can help inform whether the market is there for your product.
INTERVIEW:
ML: Thanks for joining us today, Matin. We’ll get into the more technical aspects of what Heap does later, but first, what motivated you to start Heap?
MM: Heap was born out of my experience with data analytics as a product manager for Facebook. The process of getting insights there was really, really difficult because we constantly needed to update or add tracking code to analyze data. Say I had a question about our users there was a feature I wanted to dig into further. I would need to get an engineer to write tracking code for me. Engineers usually have more important things to worry about than writing analytics code. The whole process was arduous. My co-founder and I started thinking that if Facebook couldn’t get this right despite having the best analytics infrastructure in the world, what hope did any other online business have? We started Heap to make that process much simpler. Instead of relying on others to to track, manage, and interpret data for you, we wanted to democratize access to data and bring insights to more people.
ML: So, you’ve worked at both Facebook and Google, arguably two of the most infamous players in the tech industry. What have you learned from those organizations?
MM: They were incredible companies. I learned a lot about best practices for engineering teams and how big organizations get stuff done. I also learned how dysfunctional those companies can be, especially in their approach to using data. I was convinced that there was a way to improve the feedback loop between question and answer, and that’s why Heap got started.
ML: And what are you trying to do differently than the companies where you worked before?
MM: We really believe in setting up a structure to promote individual ownership. As CEO, my role is to provide as much context as possible to everyone in the company so that they can make the right decisions autonomously. We hire smart people to tell us how to run the business, not the other way around. That said, we are still young. We are 25 people today, and we are planning on growing quickly, so what works for us now might not work when we’re 100 or 500 people. A big part of how we operate is to stay humble and pay attention to what works and what doesn’t so we can react as soon as possible.
ML: So, as a technical analytics company, I’d imagine you use Heap for your own organizational needs, right?
MM: Yes, all of us use Heap day-to-day, whether it’s for refining the product or helping us understand which of our leads are most likely to convert. I still think there is a lot we can do to improve the product, both for ourselves, current, and future customers.
ML: How would you explain to the non-technical user what Heap does and what its #1 value is?
MM: Heap takes an automatic tracking approach. Every other tool in the market takes a manual tracking approach. When you have questions about your users or your business, with Heap that data is always available to you because Heap has already collected it.You can get answers to your questions instantly without having to go through developers to write code. Heap automatically captures every user interaction like clicks, form submissions, pageviews, input changes, etc. and provides an interface for organizing and making sense of those actions after the fact.
ML: I know over 3,000 companies use Heap, but is there a certain target customer or segment that you go after that you feel can benefit the most from you?
MM: We sell to the person within an organization who is responsible for making their organization data-driven. Often that will be the Head of Analytics, the CTO, Head of Product, or an engineer. More and more companies are understanding the power of analytics these days, and it is impossible to stay relevant as a company without data driving your decisions. Not everyone is a data scientist. We want to make analytics easier for everyone within a team: marketing, sales, product, design, customer success, etc. Anyone on any team should get value out of Heap.
ML: So, one of the benefits of Heap is the speed of deployment, correct? You can start accessing the data right away? How long does it take to start using Heap Analytics? How long will it take to deploy it and start seeing the information that you need?
MM: Literally one minute. To install Heap, you add a snippet onto your site, or you add an SDK to your mobile app. As soon as it is live, we start automatically tracking user data and collect it in our system. Then, we provide a user interface for organizing and slicing and dicing that data. Let’s say I am interested how many people are clicking on a particular button to sign up for an offering. I can create that mapping in the interface, and once I have created that mapping, I can use it for ongoing analysis as if I had been tracking that event since day 1.
ML: How do you address the issue of frequent website re-design? If the website goes through weekly and monthly organizational changes, do you have to go through the process over and over?
MM: We have a first-class feature that addresses these kind of changes. We call them event combos. They let you take an old version of the event, stitch it together with a new version of the event, and then map that to a web page change or CSS change. This is then treated as one semantic entity, and Heap will just pick up on when the underlying structure of your page or app has changed.
ML: So, Heap is the only solution that allows you to do this kind of activity with instantaneous deployment. Many organizations often struggle with fully deploying Google Analytics or Adobe Web Metrics. Because Heap takes so much less time to learn and adopt, do you think programs that focus on deployment minimization will become a trend, or do you think that you might continue to be the only player in this space?
MM: We have a big head start, and I think that as our approach gains popularity, others would want to logically follow suit. At a philosophical level, the big difference is that a lot of the players in this market innovate based on the analytics and visualization layer. But the big piece we focus on (that others seem to neglect) is that having the right data is always the hardest part of analysis. It doesn’t matter how pretty your tool is or how fast it runs. If you don’t have the right data to answer your question, your analysis is fundamentally hamstrung. Thus, we focus on making sure that our customers always have the right data when they need it. As more and more people rely on data to make their decisions, we think organizations will start adopting this trend.
ML: I noticed that heapanalytics.com does not use Google Analytics, which is understandable. Right now, do you think Heap is more complementary to tools like Google Analytics, IBM, Adobe, etc.?
MM: Over time, I think Heap will eventually take more and more use cases from those tools, especially given how stagnant they’ve been. Right now, tools like Google Analytics and Omniture are strong at top-of-the-funnel /attribution analysis. For Google in particular, analytics exists to drive AdWords revenue. In terms of understanding attribution, I don’t think Heap will be able to fully corner that use case anytime soon. But I think optimizing the middle and bottom of the funnel – onboarding, retention, engagement, churn – is becoming increasingly important for businesses. Those are the kinds of things our customers need insight into.
ML: Across your 3,000 customers, how do you define the success of Heap Analytics?
MM: Customer success as a function is relatively new for us, and we’ll further consider it in 2016. We meet with our customers to make sure that they are aware of how Heap can fit their business needs. Since we capture everything, we can surface “unknown unknowns” for them in a way that no one else in the market can, and we want to make sure our customers are fully utilizing these capabilities. We also work with customers to ensure Heap serves as the backbone of their analytics strategy company-wide, not just within specific teams.. There are also a number of metrics we track within Heap itself to closely measure retention and prevent churn.
ML: As an organization that helps other companies use analytics, how do you use analytics to actually optimize Heap Analytics? What information do you collect from your customers about how they use the technology to make decisions about new tool iterations?
MM: Sure. Our product team digs into what we call the List View, where you can list individuals that fit certain criteria like whether they have signed up in the last week or how many times they have used a feature in the past month. You can look at a group of people’s behavior in a certain category, gather a list of those users and actually see common patterns throughout your site. Most product analysis starts from that macro view, and then we dig into common patterns within those groups. We use that data to inform what features to prioritize and what pieces we should re-design.
ML: So, fortunately or unfortunately, we both work in a very competitive industry. There are a lot of players to keep up with. What do you do to stay in tune with what’s happening in the field?
MM: The most important gauge for us is our customer base. Our customers will tell us what they dislike about other tools and why they’re moving over to Heap. We obviously pay attention to what is getting built and what the up-and-coming features are across the different markets. But for the most part, we have a pretty strong vision of where we want to take the company, and we rely on our customers to inform and refine that vision, rather than scrutinizing our competitors’ every move.
ML: For prospective and current users of Heap Analytics, can you share with us a couple of new capabilities that you want to launch in 2016?
MM: Absolutely! Oftentimes, companies use many different tools for analytics, depending on their needs. As the analytics market expands, tools are going to have to better integrate with each other. By virtue of capturing everything, our customers are treating Heap as their source of truth when it comes to user questions. But traditionally, Heap hasn’t made it easy to use that data within other tools. Our first step in fixing this is a feature we recently launched called Heap SQL. It helps people take the data they’ve organized in Heap and automatically keep it in sync with Amazon Redshift, which is a data warehousing tool. Heap does all of the hard ETL work of making sure that data in your Redshift cluster reflects data in Heap. Our users can then get access to the Heap data in SQL for more advanced analysis or to join it with their own data sources. We hope to expand upon Heap SQL even further in 2016.
ML: Awesome! I have one last question, not necessarily about analytics, but about entrepreneurship. Now that you’ve seen some success, do you have any recommendations for people who want to launch their own startup?
MM: Oh, boy! I think we have a lot of work to do before we can consider ourselves successful enough to give advice. That said, one of the best things an entrepreneur can do – and one of the things I was bad at as a technical person – is to sell things before they are actually built. It saves you a lot of time and can help inform whether a market exists for your product. Initially, when I first started building Heap, I was very excited about writing code and building all those features, but I think I should have talked to people more, asking them questions like, “I am going to build a solution that does this, want to buy it for $10,000 a year?” You should be asking those questions, and then see if you can get an actual commitment, if you can actually get people to put their money where their mouth is. That is the only way to validate the market.
RESOURCES:
1. Matin Movassate, the CEO of Heap – LinkedIn Profile.
2. Getting started with Heap – YouTube video.
3. Heap website and @heap Twitter page.
